Clinical validation of an algorithm to correct the error in the keratometric estimation of corneal power in normal eyes.

Purpose: To validate clinically in a normal healthy population an algorithm to correct the error in the keratometric estimation of corneal power based on the use of a variable keratometric index of refraction (n(k)).

Setting: Medimar International Hospital (Oftalmar) and University of Alicante, Alicante, Spain.

Design: Case series.

Methods: Corneal power was measured with a Scheimpflug photography-based system (Pentacam software version 1.14r01) in healthy eyes with no previous ocular surgery. In all cases, keratometric corneal power was also estimated using an adjusted value of n(k) that is dependent on the anterior corneal radius (r(1c)) as follows: n(kadj) = -0.0064286 r(1c) +1.37688. Agreement between the Gaussian (P(c)(Gauss)) and adjusted keratometric (P(kadj)) corneal power values was evaluated.

Results: The study evaluated 92 eyes (92 patients; age range 15 to 64 years). The mean difference between P(c)(Gauss) and P(kadj) was -0.02 diopter (D) ± 0.22 (SD) (P=.43). A very strong, statistically significant correlation was found between both corneal powers (r = .994, P<.01). The range of agreement between P(c)(Gauss) and P(kadj) was 0.44 D, with limits of agreement of -0.46 and +0.42 D. In addition, a very strong, statistically significant correlation of the difference between P(c)(Gauss) and P(kadj) and the posterior corneal radius was found (r = 0.96, P<.01).

Conclusion: The imprecision in the calculation of corneal power using keratometric estimation can be minimized in clinical practice by using a variable keratometric index that depends on the radius of the anterior corneal surface.

Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned.